Number
826
Name
Tailored GenAI podcasts as High-Yield Pre-Reading for Foundational Sciences in Early Medical Education.
Date & Time
Monday, June 8, 2026, 6:00 PM - 7:30 PM
Location Name
Oglethorpe Ballroom
Speakers
Authors
Renu Agnihotri, ATSU
Presentation Topic(s)
Technology and Innovation
Description
PURPOSE
Medical and dental basic science students at ATSU identified a critical
need for focused, digestible content that synthesizes information and
provides a high-yield overview prior to lectures. The need was especially
acute for complex, detail-intensive subjects like medical genetics. This
project aimed to leverage emergent GenAI tools to create an efficient,
accessible method for synthesizing complex information to meet this
instructional demand.
METHODS
The project utilizes the institutional access to the secure Google for
Education suite, specifically Notebook LM, to generate tailored audio guides
from selected text sources. For complex topics, these critical sources
include specific textbook chapters, lecture slides, and board-focused
materials, all carefully aligned with session learning objectives to ensure
rigor and robust content. The guides are structured as conversational
podcasts, typically running approximately 20 minutes, simulating a relaxed
conversation with natural, dynamic voices. These source-based podcasts are
applied as pre-read tools across all sessions in genetics within the basic
science coursework in Medical and Dental School Years 1-2. Additionally
students complete self-assessment concept checks (practice questions) tightly
aligned with learning objectives.
RESULTS
The AI-generated podcasts are currently being implemented across the basic
science curricula for both medical and dental students this term. Initial
student feedback has been reported as very positive and enthusiastic,
confirming the tool's high utility for synthesizing basic science
information. Student feedback, including qualitative reports and quantitative
usage data, will be collected at the end of the term to fully assess the
tool's impact.
CONCLUSION
The development and deployment of customized, source-based AI podcasts
effectively meet the identified student need. This approach provides a
reliable and scalable model for ethically integrating AI into medical
education and aligns with ATSU's institutional goals. Future steps involve
scaling up the innovation and offering personalized training.
Medical and dental basic science students at ATSU identified a critical
need for focused, digestible content that synthesizes information and
provides a high-yield overview prior to lectures. The need was especially
acute for complex, detail-intensive subjects like medical genetics. This
project aimed to leverage emergent GenAI tools to create an efficient,
accessible method for synthesizing complex information to meet this
instructional demand.
METHODS
The project utilizes the institutional access to the secure Google for
Education suite, specifically Notebook LM, to generate tailored audio guides
from selected text sources. For complex topics, these critical sources
include specific textbook chapters, lecture slides, and board-focused
materials, all carefully aligned with session learning objectives to ensure
rigor and robust content. The guides are structured as conversational
podcasts, typically running approximately 20 minutes, simulating a relaxed
conversation with natural, dynamic voices. These source-based podcasts are
applied as pre-read tools across all sessions in genetics within the basic
science coursework in Medical and Dental School Years 1-2. Additionally
students complete self-assessment concept checks (practice questions) tightly
aligned with learning objectives.
RESULTS
The AI-generated podcasts are currently being implemented across the basic
science curricula for both medical and dental students this term. Initial
student feedback has been reported as very positive and enthusiastic,
confirming the tool's high utility for synthesizing basic science
information. Student feedback, including qualitative reports and quantitative
usage data, will be collected at the end of the term to fully assess the
tool's impact.
CONCLUSION
The development and deployment of customized, source-based AI podcasts
effectively meet the identified student need. This approach provides a
reliable and scalable model for ethically integrating AI into medical
education and aligns with ATSU's institutional goals. Future steps involve
scaling up the innovation and offering personalized training.